On Wed, Dec 13, 2017 at 08:50:33AM +0100, Vladimir Vassilev wrote: > > At the current point I think all issues raised are addressed with the > following model (notice the added anydata option which can replace the use > of yang-library-datastore and if implemented as an alternative can make > retrieval as a single tree): > > module: ietf-datastores > +--ro datastores-state > +--ro datastore* [name] > +--ro module-set-id string > +--ro (model)? > +--:(same-as-operational) > +--:(constrained-to-operational) > | +--ro not-implemented* [name revision] > | +--ro name -> /yanglib:module-state/module/name > | +--ro revision -> /yanglib:module-state/module/revision > +--:(unconstrained-datastore-instance) > | +--ro yang-library-datastore identityref > +--:(unconstrained-anydata) > +--ro yang-library? <anydata> > > augment /yanglib:yang-library-change: > +---- datastore? identityref >
It is (a) architecturally backwards to have ietf-datastores depend on ietf-yang-library and (b) the proposal does not solve the issue that you are silently changing the semantics of the definitions in the old ietf-yang-library. It remains unclear to me which problem this is approach solving, i.e., what the benfit is over the other proposals (since the semantics of ietf-yang-library change, it is _not_ backwards compatible). /js -- Juergen Schoenwaelder Jacobs University Bremen gGmbH Phone: +49 421 200 3587 Campus Ring 1 | 28759 Bremen | Germany Fax: +49 421 200 3103 <http://www.jacobs-university.de/> _______________________________________________ netmod mailing list [email protected] https://www.ietf.org/mailman/listinfo/netmod
